Roger E. A. Arndt, Principal Investigator

Unsteady Sheet/Cloud Cavitation

Ventilated test body.

Cavitation is a significant consideration in a variety of important engineering applications. As the performance of pumps and turbines is increased and hydrofoil ships are designed for higher speeds, it becomes necessary to design lifting surfaces that can operate effectively in the cavitating mode.

This group’s research indicated that sheet/cloud cavitation is a complex and very substantial subset of the overall problem. Focusing on the strongly periodic formation of cloud cavitation that leads to a structured wake consisting of vortical clouds of bubbles, the group found that the phenomenon leads to unsteady lift, which cannot be accurately predicted at this stage. Therefore, the aim was to develop the computational tools to study this problem. Such computational efforts were supplemented by detailed experimental work that is currently underway.

Preliminary computation.

This work included an exploration of the commercial software package fluent 5’s ability to capture the dynamics of cavitation. The numerical approach used by fluent 5 is part of the Rayleigh-Plesset equation family of cavitation models. This model was implemented along with the Volume of Fluid (VOF) method to track the void fraction in the flow. fluent 5 can calculate the cavitating flow in both threedimensional and two-dimensional grid systems. The main investigation was performed on a two-dimensional NACA 0015 hydrofoil grid, designed to resemble a test tunnel setup located at St. Anthony Falls Laboratory (SAFL). The study was performed for cavitation numbers ranging from inception to super-cavitation at two different angles of attack. All numerical results were compared quantitatively to corresponding experimental results from the SAFL tunnel. This phase of the study was carried out using the resources made available at the Supercomputing Institute.

A further investigation of the three-dimensional model was performed, using the grid of a low head Francis hydro-turbine. However, comparison data were not available for this case, a situation that leads to a more qualitative approach than in the two-dimensional case. The numerical investigation of the hydro-turbine was only performed at best efficiency operation. As such, this research showed that the fluent solver can capture several features of cavitating flows, and can probably be used for industrial purposes.

As an extension of their work on sheet cloud cavitation, this group has broadened their simulations to include ventilated flows and outgassing due to cavitation. This is important because of the industrial applications of cavitation. The existing model, Arndt et al. (2000), has been modified to include a method for taking into account incondensable gas. Previously, the model only considered condensable vapor. In order to develop and test the model, the group selected ventilated flow as a test case. Ventilation is used in applications of cavity dynamics by injecting air into a region of low pressure, allowing supercavitation to develop at higher values of cavitation number (e.g. free stream pressure). An important question is the amount of ventilated flow required. This question can only be answered if a complete understanding of the physics associated with the bubbly wake trailing such a body is known.

Research Group and Collaborator

Gotfred Berntsen, Norges Teknisk-Naturvitenskapelige Universitet (NTNU), Trondheim, Norway
Morten Kjeldsen, Research Associate
Michael Levy, Research Associate
Quao Qin, Graduate Student Researcher
Travis Schauer, Graduate Student Researcher